initial light and dynamic convolution kernels (#547)

Summary:
CUDA code for light/dynamicconv kernels, including pytorch modules. Modules can be built by running setup.py in each respective folder, and can then be imported and used like any other module.
Pull Request resolved: https://github.com/fairinternal/fairseq-py/pull/547

Reviewed By: myleott, shubho

Differential Revision: D15703660

Pulled By: nng555

fbshipit-source-id: e9c913753be3a1cd571965f7200df6678b644520

fairseq/modules/lightconv_layer/setup.py

+from setuptools import setup
+from torch.utils.cpp_extension import CUDAExtension, BuildExtension
+
+setup(
+    name='lightconv_layer',
+    ext_modules=[
+        CUDAExtension('lightconv_cuda', [
+            'lightconv_cuda.cpp',
+            'lightconv_cuda_kernel.cu',
+        ]),
+    ],
+    cmdclass={
+        'build_ext': BuildExtension
+    })

fairseq/modules/lightweight_convolution.py

@@ -10,6 +10,21 @@ import torch.nn.functional as F
 from fairseq import utils
 from fairseq.modules.unfold import unfold1d
 
+def LightweightConv(input_size, kernel_size=1, padding_l=None, num_heads=1,
+                    weight_dropout=0., weight_softmax=False, bias=False):
+    if torch.cuda.is_available():
+        try:
+            from fairseq.modules.lightconv_layer import LightconvLayer
+            return LightconvLayer(input_size, kernel_size=kernel_size,
+                                  padding_l=padding_l, num_heads=num_heads,
+                                  weight_dropout=weight_dropout,
+                                  weight_softmax=weight_softmax, bias=bias)
+        except ImportError as e:
+            print(e)
+    return LightweightConv1dTBC(input_size, kernel_size=kernel_size,
+                                padding_l=padding_l, num_heads=num_heads,
+                                weight_dropout=weight_dropout,
+                                weight_softmax=weight_softmax, bias=bias)
 
 class LightweightConv1d(nn.Module):
     '''Lightweight Convolution assuming the input is BxCxT

fairseq/modules/unfold.py

@@ -5,7 +5,6 @@
 
 import torch.nn.functional as F
 
-
 def unfold1d(x, kernel_size, padding_l, pad_value=0):
     '''unfold T x B x C to T x B x C x K'''
     if kernel_size > 1: